System and method for labeling a document
AssigneeYahoo! Inc.
InventorsWinton Davies, Deepa Joshi
Life of this patent
Estimating the term…
Abstract
A system and method for selecting a proxy keyword for an unknown document. An unknown document is received by a receiver. A plurality of candidate documents and corresponding keywords are determined for the unknown document. Using the keywords from the candidate documents, proxy keywords are determined for the unknown document based on a plurality of factors including a length of the keywords, a distance of the candidate documents from the unknown document, a similarity of the text between the unknown document and the respective candidate document, a rank of the keywords within each candidate document, and a frequency of the keyword within its respective candidate document.
Claims21
A computerized method for selecting a proxy keyword for a document, the method comprising:
- receiving an unknown document;
- electronically determining, using a processing device, at least a first candidate document from the world wide web for the unknown document on the basis of a search of one or more valued search terms;
- determining at least a first keyword for the first candidate document;
- electronically determining, using the processing device, at least a second candidate document from the world wide web for the unknown document on the basis of a search of one or more valued search terms;
- determining at least a second keyword for the second candidate document;
- electronically determining, using the processing device, a proxy keyword for the unknown document based on the first and second keywords;
- and
- wherein the first candidate document and the second candidate document are documents which are most similar to the unknown document.
A system for selecting a proxy keyword for a document, the system comprising:
- a receiver effective to receive an unknown document;
- a candidate document determination module effective to determine at least a first candidate document from the World Wide Web for the unknown document on the basis of a search of one or more valued search terms;
- a proxy phrase determination module effective to determine at least a first keyword for the first candidate document;
- the candidate document determination module further effective to determine at least a second candidate document from the World Wide Web for the unknown document on the basis of a search of one or more valued search terms;
- the proxy phrase determination module effective to determine at least a second keyword for the second candidate document;
- and
- a processor effective to determine a proxy keyword for the unknown document based on the first and second keywords;
- and
- wherein the first candidate document and the second candidate document are documents which are most similar to the unknown document.
Dependencies are inferred from the claim text.